运动人体识别技术研究.docVIP

PAGE PAGE I 运动人体识别技术研究 摘要 随着计算机视觉、模式识别等相关技术的发展，基于视频的运动人体检测和跟踪受到了多方关注，被广泛应用到银行、商场、停车场等对安全要求较高的地域监视中。基于视频的方法是唯一一种无入侵的运动捕获方法在海量视频数据检索、人机交互、视频监控、体育运动分析等领域有广泛的应用前景。工作目的：运动人体识别的研究。研究方法：分析比较当前社会下人体识别的不同类型以及优缺点。成果和结论：面对着全球化、信息化越来越提倡的社会背景下，识别技术会越来越广，深度也会不断加深。其虽然有着自己独特的优势，但并不是没有缺点。实时场景下的人体跟踪会比较复杂，存在背景变化、遮挡、噪声及其他干扰因素，难以同时满足准确性、鲁棒性和实时性的要求。本文对单个摄像机且摄像机静止条件下的人体检测跟踪进行了研究，分析了现在已知的一些检测跟踪方法，采用背景减除法和投影比例检测法，提出了一种适用于室内室外环境，由运动物体检测、运动物体分析和人体运动跟踪3部分组成，使用单个静止数字摄像机得到德尔灰度图像序列作为输入，能够对单人或多人的运动行为进行监控的视频检测与跟踪系统。 关键词：人体跟踪；人体检测；运动人体； detection and tracking of human according to the feature of human ABSTRACT With the development of the related technology of computer vision, pattern recognition, human motion detecting and tracking based on the video received much attention, and has been widely applied to banks, shopping malls, parking and other regional monitoring of higher security requirements of. Video based method is only a field without invasion of the motion capture method in video data retrieval, human-computer interaction, video surveillance, sports analysis has a wide application prospect. Job purpose: Research on moving human body recognition. Research methods: different types of analysis of human recognition compared to current society and the advantages and disadvantages. Results and conclusion: in the face of globalization, informatization is increasingly advocated social background, recognition technology will be more and more widely, will also continue to deepen the depth. Although it has its own unique advantages, but not without disadvantages.Real time scene human tracking can be complicated, the presence of background changes, occlusion, noise and other interference factors, it is difficult to meet the accuracy, robustness and real-time requirements at the same time. This paper studied the human detection and tracking of single camera and the camera under static condition, analyses the some known detection and trackin